Language for work and work for language: linguistic aspirations in the marketing of domestic work

1 Introduction

Commodified domestic work, that is, economic transactions in exchange for home-cleaning and other domestic services, is a global labor market characterized by migration (Gonçalves and Schluter 2020; Lutz 2011). One in six domestic workers worldwide are transnational migrants (ILO 2015). In Sweden, cleaner is the most common occupation among migrants, and, furthermore, migrants account for two out of three workers in the sector “other home-based personal care” (Statistics Sweden 2020). While this connection between migration and domestic work has become stronger in the last couple of decades, the connection is salient throughout the twentieth century in Sweden.

The connections between transnational migration and domestic work have made the sector an interesting focus for sociolinguistic inquiry. Our aim in the present study is to shed light on the ways in which individuals and companies refer to second language speakership in marketing for domestic services. Additionally, the study addresses how these references can serve to navigate tensions – or ‘spaces of upset’ (see Introduction this issue) – associated with migration and commodified domestic work.

Transnational migration tends to engender upset in various ways, and language is often in the line of fire. Adult migrants’ (perceived) lack of language competence is often perceived as a threat to social cohesion and security (Kahn 2019; Kraft 2019). Simultaneously, the individual’s language learning is constructed as a moral obligation linked to non-linguistic characteristics of diligence and belonging in the new country of residence (Allan 2013; Heller and Duchêne 2012; Kahn 2019). As domestic work has been conceived as manual low-skilled work which does not require advanced language skills, it is seen as a viable entry point into the labor market for migrants (Strömmer 2020). With respect to employability, speakers’ multilingual repertoires are often evaluated in relation to the local linguistic market (Flubacher et al. 2018). The language competence of adult migrants is connected to work in at least two recurring language ideologies: one positing language competence as a prerequisite for work (language for work) and another positing work as a key site for language learning (work for language). Both of these ideas are salient in connection to domestic work. The ‘language for work’ idea is exemplified in language training courses for immigrants that are specifically targeted at domestic work, and in language requirements in recruitment processes where a certain level of language competence is turned into a minimum requirement (Lorente 2012, 2018; Strömmer 2020). A context where the ‘work for language’ idea is made relevant is in political discourses on the workplace as a site for language learning and integration (Lindberg and Sandwall 2017). ‘Work for language’ is also relevant in the au pair system, a form of linguistic and cultural exchange where a usually young person from another country comes to live in a family to do lighter domestic services, including childcare.

In addition to the morally loaded values connected to second language speakership, domestic work is, in and of itself, a space of upset marked by tensions concerning social inequalities, precarious working conditions, and conflicts of interest between employer and employee. Research on language and transnational domestic workers has delved into the ways in which the relationship between employer and employee, in these cases, is often shaped by an uneven distribution of linguistic resources. For instance, in a study of Filipino domestic workers in Singapore, Lorente (2018) illustrates how both language competence and ethnicity are used in the construction of hierarchies between domestic worker and employer. In Ladegaard’s study of migrant domestic workers in Hong Kong (2020), the domestic workers’ lack of proficiency in the language of the employer aggravated a situation of abuse. In the context of commodification of domestic work, Sweden represents an interesting case. The market of domestic services has been met by a displayed ambivalence, as this phenomenon is conceived as materializing stratified and racialized social relations that clash with a self-image of Sweden as a just and equal welfare state (Platzer 2006; Widding Isaksen 2010).

This article is organized as follows: First, we give a brief overview of the changing conditions of migration and domestic work in Sweden. This is followed by a presentation of the study’s analytical framework. We then present the data drawn on in the study: advertisements for domestic work, offered first by individual domestic workers in the mid-twentieth century, and later by cleaning agencies in the early twenty first century. The analysis is presented in Sections 5 and 6, starting with the dataset of historical personal advertisements, and ending with the dataset of contemporary marketing from cleaning agencies. In the concluding remarks, we sum up our argument that referring to language learning aspirations and language competence can be a discursive strategy to attribute positive values to matters laden with unease. In our data, to talk about language learning seems to be a way to navigate the space of upset constituted by the phenomenon of migratory domestic work.

2 Migrant domestic work as a space of upset and as a problem solver

Contemporarily as well as historically, commodified domestic work has been characterized by socio-economic differences between employer and employee. The general pattern is that middle- and upper-class families buy the services from less privileged individuals (Gonçalves and Schluter 2020), and the sector has long (or always) constituted a gendered and racialized labor market (Parreñas 2015). Materializing social and economic inequalities in Swedish society, commodified domestic work has repeatedly engendered national debates of upset regarding socio-politics and morality.

Historically, having domestic help was an important aspect of a bourgeois lifestyle in Sweden. From the 1930s onwards, hiring domestic help became a solution for the growing number of middle-class women who wanted to have both a family and a career. This was also encouraged politically as a way to increase birth rates and enhance the socio-economic status of middle-class women (Platzer 2006). The demand for domestic workers increased substantially in the first decades of the twentieth century in Sweden. However, the number of individuals who were willing to work in private households decreased, as the occupation came to be perceived as unmodern and associated with poor working conditions, and as the labor market for women expanded with other options. The subsequent so-called “servant crisis”, with unmet demands for domestic help, had its most significant phase between the 1930s and the 1960s. The “servant crisis” was considered a social problem and was offered some political attention at the time (Nordlund Edvinsson and Söderberg 2010; Platzer 2006).

As domestic work blurs the line between work and private space, there is a risk of exploitation associated with the occupation (Lønsmann 2020). However, with respect to the needs of the employers and the perceived nature of the work, domestic labor remained unregulated for a long time (Calleman 2011). Documents such as letters, diaries, and autobiographies from the first half of the twentieth century point to the hardships of domestic work (Nordlund Edvinsson and Söderberg 2010). They show how the relationship between employer and employee embodied class and gender conflicts and was sometimes marked by frictions. In an effort to make the occupation more attractive, the so-called servant crisis was met by a state commission report and a law that regulated the domestic workers’ working hours (Hembiträdeslag [‘Act on Housemaids’], government bill 1944: 217). Another effort to solve the servant crisis was to make it easier to recruit migrant labor (Anving and Eldén 2016). One attempt to this end was to exempt domestic labor from work permit requirements. Furthermore, refugee women from the non-Nordic countries who sought employment in Sweden were consigned to domestic work for two years before they could seek other employment (Strollo 2013). Thus, the maids recruited from abroad provide an example of how migration was used to manage the local labor market from the perspective of prospective employers. Indeed, a large proportion of work immigrants in the 1950s were female domestic workers, with the most common nationalities being German, Finnish and Norwegian.^[1] Meanwhile, these female migrant domestic workers and their working conditions were largely invisible in the public debate at the time (Strollo 2013). This invisibility is also reflected in research, as the sociolinguistic realities of transnational domestic workers have mainly been studied from a contemporary perspective (e.g. Ladegaard 2020; Lorente 2018; Lutz 2011; but see Divita [2020] for an exception). Another lucid exception is Strollo’s retrospective interview study (2013), which investigates the experiences of German women who came to Sweden in the 1950s for jobs in the domestic sector. Strollo gives examples of how German maids were expected to teach German to the children of the house and shows how they saw their domestic employment as a way to migrate to Sweden, even though they considered the occupation unmodern and temporary.

Swedish contemporary history is marked by the fact that the Social-Democratic Party headed the government continuously for four decades, from 1936 to 1976. During this period, there was a significant extension of the Swedish welfare system, in which the provision of care services is a cornerstone (Widding Isaksson 2010). A general discourse on the importance of equality, as well as the provision of state-subsidized childcare and parental leave, led to a decreasing demand for domestic help. This in turn led to domestic workers being almost absent from the labor market in the 1970s (Platzer 2006). However, the market for domestic work re-emerged in the 1990s (Anving and Eldén 2016; Calleman 2011). This development has been explained with reference to a number of societal changes, including growing unemployment, continuous immigration, the enlargement of the EU, increasing wage differences, increasing numbers of dual career families, and cuts in the public sector (Calleman 2011; Gavanas 2006; Platzer 2006). Taken together, societal changes in the latter half of the twentieth century generated a situation where more people could afford to pay for domestic services at the same time that more people were willing to work in the sector. Since the 1990s, the demand for domestic work in private households has been steadily increasing.

Meanwhile, when the market for commodified domestic services re-emerged, there was a significant difference in terms of how the transactions were organized: from in-house maids to cleaning agencies providing domestic services by the hour. Unlike countries in other parts of the world (cf. Lorente 2018), domestic services in Sweden today are mainly provided by a cleaner from a cleaning company, who comes to private households during the day when no one is at home. The proliferation of cleaning agencies is largely due to the so-called RUT deduction,^[2] a tax deduction that effectively halves the price for domestic services. The idea of the RUT deduction was launched in 1993 as an attempt to ease the work-life balance in dual career families and to reduce the market of undeclared domestic work. It was put into place in 2007, after 14 years of heated debate on the deduction itself, but also on the very idea of selling and buying domestic services. The debate went under the popular name pigdebatten (‘the maid debate’), using the word piga, an older word for female servants in rural contexts, to project the idea that commodified domestic work symbolized a regression to an outdated, unjust society, which clashed with the idea of Sweden as a leading country with respect to gender and social equity (Gavanas 2006; Widding Isaksen 2010). In this vein, the debate was based on an interdiscursive link to a particular idea of historical time – and a type of character therein – materialized in newspaper headlines like “The maid resurrects as an entrepreneur” (Svenska Dagbladet, November 8, 1996). Increasingly, RUT was also perceived as a solution to unemployment problems, in particular with respect to foreign-born Swedes positioned far from the labor market. As the profiled union leader Björn Rosengren, later to become Social-Democratic minister of enterprise, proposed: “Let immigrants clean the homes!” (Dagens Nyheter, 17 December 1995).

Commodified domestic services have simultaneously been presented as a problem solver on a personal level. An interview study with dual career middle-class families demonstrates that a major reason for buying domestic services is to avoid conflicts between husband and wife (Platzer 2006: 216). Similarly, Eldén and Anving (2016) show how the rationale for recruiting an au pair is to solve the work-life balance and gender equality in dual career families. The family ease expected to arise from buying domestic services is indexed in the names of some contemporary cleaning agencies, such as Hemfrid and Vardagsfrid (‘home peace’, ‘everyday peace’).

Hence, there is a double indexicality linked to commodified domestic services. On the one hand, it is seen as a problem solver on a personal and societal level: a facilitator for achieving a modern gender equality ideal and an opportunity for the economic integration of migrants. On the other hand, domestic work is met with unease as it is seen as reminiscent of an outdated social order and of a socially and racially stratified society. The domestic services market thus seems to remind the civic conscience of a social hierarchy that most would prefer to overlook. Hence, historically as well as contemporarily, domestic work has been a space of upset.

3 Advertising domestic work – a framework for analysis

The present study explores how references to language are used in the marketing of transnational domestic workers in order attain different kinds of symbolic values. Following Heller and Duchêne (2012), we will analyze the ways in which language becomes a commodified resource associated with symbolic value, which regulates access to other resources. More specifically, the analysis centers on the ways in which language competence and language learning aspirations are drawn on as symbolic capital in advertisements concerning domestic work (Bourdieu 1991).

Advertisements are conditioned by spatial restrictions, and the phrasings need to be carefully selected. While the author does not know who the readers will be, it is reasonable to assume that each advertisement is written with a generalized reader in mind; a prospective buyer of the services offered. In this vein, the phrasings in the advertisements can be understood as semiotic work designed in anticipation of others’ interpretations of one’s self-presentation (Visakko 2015). Writing an advertisement can be understood as an attempt to market an object as favorably as possible with respect to an anticipated profit (here, being offered a position, or contracting more customers). Meanwhile, neither those who wrote the advertisements, nor us as analysts, can say whether the advertisements were successful in this respect.

Advertisements can be understood as a mediated genre, referring to “a constellation of systematically related, co-occurrent formal features and structures that serves as a conventionalized orienting framework for the production and reception of discourse” (Bauman 2004: 3–4). Hence, we recognize a genre partly due to recontextualized linguistic signs, such as genre-specific phrasings, often in relation to designated spaces (such as a particular section in a newspaper or a billboard). Thus, a specific advertisement is intertextually linked to other advertisements. However, as Bauman goes on to argue, language use is conditioned by an intertextual gap, and, therefore, the conventionalized organization of discourse is not sufficient in its capacity to account for specific uses of discourse. In other words, there is always space for creativity. Indeed, part of the advertisement logic is the need to be noticed and remembered among other advertisements.

Furthermore, the advertisements under investigation can be understood as a materialization of a linguistic market. Bourdieu (1991: 77) underscores that “the constraint exercised by the market via the anticipation of possible profit naturally takes the form of an anticipated censorship, of a self-selection which determines not only the manner of saying, that is the choice of language […] but also what it will be possible to say.” In the line of this argument, Bourdieu posits that individuals have what he calls a ‘linguistic sense of placement’. Developing this concept further in a sociolinguistic context, Salö (2015) defined this as “a fine-tuned ability to sense and anticipate the valorization of linguistic goods in different markets, which reflexively allows agents to align their linguistic practices in relation to present and upcoming exchanges” (Salö 2015: 517). Following this line of thought, references to language learning and language competence in advertisements for domestic work can be seen as metalinguistic products that reflect the understanding that the author of each advertisement had of the linguistic market and the place that various speakers would be allotted therein.

4 Data and method

The study at hand draws on two sets of data: one historical and one contemporary. The historical dataset is a collection of personal advertisements by job seekers, published 1905–1991 in the Swedish daily Svenska Dagbladet, where the phrase lära sig svenska (‘to learn Swedish’) is used in advertisements for domestic positions. The archive contains 188 personal advertisements by job seekers where the phrase is used, and the 119 of these advertisements that directly target jobs in the domestic sector constitute our dataset.

The advertisements do not provide much information about the applicants, although we can learn that all but two are women. Furthermore, most applicants state their nationality, so we know that the most common nationalities are Finnish (N = 70) and German (N = 20). Additionally, the following nationalities figure once or a few times each: American, Austrian, Dutch, English, Estonian, French, New Zealand, and Swiss. Virtually all advertisements in the newspaper are written in standard, native-like Swedish, and given that the applicants state that they wish to learn Swedish, it is likely that someone else wrote the advertisements. Some advertisements explicitly indicate that they are written by an advertising agency either in Sweden or abroad.

The contemporary dataset is from 2018 to 2020 and consists of commercial marketing by agencies that provide domestic services. This dataset was created through purposive sampling online, in urban spaces, and in print. The majority of it was produced by looking through the websites of 50 randomly sampled home-cleaning companies in Stockholm, using an online directory of businesses. From these websites, examples were collected of discourse about language competence and language learning in relation to the workforce. Additionally, occasional public marketing which has suited the purpose of the study has also been included in the data collection. In sum, the study has thus included discourse about language and domestic work from seven different companies.

Although the two datasets differ, what they have in common is that they are forms of marketing that address prospective buyers of domestic services and that they can be analyzed as semiotic behavior aiming to promote oneself, either as an individual or as a company. Important differences include who is displayed as the sender, and the visibility of those performing the domestic services. In the historical dataset, the advertisements portray the individual domestic worker, while in the contemporary dataset, the domestic workers are present, but not primarily as individuals. Rather, they form part of the package of commodified domestic services mediated by cleaning agencies.

Analyzing these data, the focus is on (1) the ways in which references are made to language learning and language competence in the marketing of individual job seekers and cleaning companies and (2) how these references can be understood. The analysis, presented below, is divided into two parts, according to the two datasets.

5 Language learning aspirations in personal job advertisements

This section presents an analysis of the ways in which the phrase “to learn Swedish” figures in personal advertisements by job seekers in a Swedish daily 1905–1991. The majority of the investigated advertisements were published between 1930 and 1960, thereby coinciding with the most significant phase of the so-called servant crisis (Nordlund Edvinsson and Söderberg 2010), as well as with the point in time when Sweden’s immigration for the first time outnumbered its emigration.

What the advertisements have in common is the expression of a dual objective: they express a wish to find a domestic position, as well as a wish to learn Swedish. More particularly, all of the advertisements contain the exact phrase lära sig svenska (‘to learn Swedish’). The description of the kind of position sought or the services offered is framed in various and sometimes rather vague ways. Some use generalized descriptions like the title hembiträde (‘maid’) or the phrase gå frun tillhanda (‘assist the lady of the house’), or list possible household chores.

The phrase “to learn Swedish” is used in different ways in our collection. The most common is to place the phrase in the descriptive text, but there are also a few advertisements where the phrase is used as a heading or as a signature. Four examples are presented below.

Example 1.

April 17, 1909

In order to learn the Swedish language

a young, educated German wishes to find a position in a family as company and help for the mistress in the house in exchange pocket money. Excellent references. Response marked […]

Example 2.

July 22, 1939

21-year-old German-speaking girl

linguistically knowledgeable, who wishes to learn Swedish, is seeking placement in good family with cultural interests, preferably where music is practiced. Willing to assist the lady of the house and look after children. References are wanted and given. […]

Example 3.

August 28, 1948

Maid,

Dutch, 25 years, able in household chores, w. to come as a fam. memb. to a Sw. fam., to learn Swedish. Speaks Eng. and German. Resp. in Eng. to […]

Example 4.

October 11, 1951

Two middle school experienced

19–20-year-old prudent Finnish girls w. f. pl. [wish for placement] in a good home in Sweden (together or apart). Childcare or work of any kind. What we do not know we wish to learn. Respond to “In order to learn Swedish” […]

6 Language learning and language competence in contemporary commercial marketing

In twenty first century Sweden, the demand for domestic services is no longer met by live-in maids, but more often either by agencies who sell cleaning services by the hour or by non-Swedish youth under an au pair contract. The re-emerging demand for domestic services, together with the introduction of the RUT tax deduction system, have led to a surging market for companies that sell hour-based domestic services such as laundry, babysitting, and, most commonly, home-cleaning. This new, competitive market for cleaning companies has engendered a need for distinctive marketing. This section presents an investigation of the ways that language competence and language learning figure in the marketing of contemporary cleaning agencies, either in public advertisements or on their webpages. The investigation carried out in this study indicates that it is far more common for cleaning companies to not mention anything about language in their marketing. Yet, the ways in which language does occasionally occur in such marketing is worthwhile probing more deeply in order to grasp the role that second language speakership has in the space of upset constituted by migratory domestic labor.

One interesting tendency is the way in which references to the language competence of the workforce is mixed up with some form of ethical profiling. Example 5 shows an advertisement from a campaign published in a magazine, but the advertisement also appeared on public billboards in the Stockholm area.

Example 5.

Ad in Hemma i HSB, May 2018

Translation:

Clean with clean conscience

(…)

✔ Collective agreement

✔ A regular cleaner

✔ We speak Swedish

Under the heading “Clean with clean conscience”, the advertisement lists and ticks “collective agreement”, “a regular cleaner”, and “we speak Swedish”. Framing the service as a commodity that can be bought “with clean conscience”, the advertisement seemingly alludes to previous debates about undeclared work and precarious working conditions among domestic workers, pre-empting the potential worries of the addressee by ticking the “collective agreement” box. Subsequently, the well-known fact that a large share of cleaners are new residents in the country is connected to this discourse of social justice. The fact that the company (“we”) can speak Swedish is, apparently, a reason for the potential customer to have a clean conscience. The language competence of the workforce is used, together with the promise of a regular cleaner and collective agreement coverage, in the company’s displayed quality benchmark.

A similar discursive strategy is found in the marketing of another cleaning agency, as shown in Example 6.

This company connects language competence to another mark of quality and safety, namely, the promise of a regular cleaner. In other parts of the same text, not recounted in the extract, the company rhetorically asks, “Why should you hire us?” and goes on to presents itself as a “serious home cleaning company”, declaring, among other things, that they follow a collective agreement contract. Notably, when the language competence of the cleaners is in focus, this company talks about itself as a “we” that does not include the cleaners, but which is rather in possession of resources in the form of cleaners. The fact that the company has high demands on the Swedish skills of “their” cleaners is presented as a mark of quality. In this sense, the marketing consists of first-language users who represent the company talking about second language users who work for the company.

Other instances of self-promotion can rather be seen as a concession to a multilingual reality. One company (the same as in Example 5) has a section on its website that lists reasons for choosing them. One of the six reasons listed has the heading, in bold, “Swedish-speaking” – but the descriptive text acknowledges the diverse linguistic repertoires of the workforce included in the company “we”.

Even though the multilingual resources of the workforce are acknowledged, a linguistic hierarchy is constructed between Swedish (the language of staff management) and competence in a large number of unnamed languages. This linguistic hierarchy applies to employees at the office and those employees who perform the domestic services. Customers are, tongue-in-cheek, promised the opportunity to “talk poetry with the office”, but are asked to speak more slowly “in the field” (that is, in their homes, when they occasionally meet a cleaner there). It is apparently taken for granted that customers will want to use Swedish, and that they will need to pace themselves in order to speak a form of Swedish simple enough for the cleaners to understand. Arguably, even though this instance of corporate self-promotion alludes to the multilingual resources of the workforce, the remaining argument is that the staff speaks Swedish and that this is a mark of quality (it is, after all, listed as a reason for choosing the company).

Another way of referring to the second language speakership of the workforce is found in one of Sweden’s largest companies in the sector. This company also has a section on its website that lists reasons for choosing them instead of others, and language turns up here, too.

Again, language figures in a discourse of quality, labor rights, security and trustworthiness. In addition to the fact that the staff is trained and covered by a collective agreement, the workforce is also offered language tuition. The fact that this company offers its employees language teaching at work is recurrently addressed in their marketing. In 2019, the company invested in an advertisement campaign in urban areas in Stockholm, which, interestingly, did not address customers, but prospective employees.

Translation: [Second from left] “Get a first job and a second language.” [Middle] “With us, you get secure employment and the opportunity to study Swedish at work. Check out our vacancies at hemfrid.se.” Commercial advertisement in Gärdet Metro station. Stockholm. (Photograph by second author, September 18, 2019)

The campaign presents employment in the domestic sector as a viable pathway to integration and to learning Swedish (see also Leppänen and Westinen this issue). Here, the prospective employee is tempted with the offer to “get a first job and a second language”, and is explicitly promised the opportunity to study Swedish at work. It is, arguably, not economically feasible for the company to invest in a campaign like this only to get more people to work for it. Instead, it is reasonable to believe that prospective customers are implicitly targeted as well, and the message to them is clear: hiring us is nothing at all like supporting a regression to an outdated, unjust society, but is rather a good deed in favor of integration and language learning. Another example is found in a corporate narrative of a small cleaning company.

This narrative tells the story of the commencement of the company. Here, offering a position to a migrant with little competence in Swedish is presented as having a larger benefit for the employee than the employer. As the story goes, one of the few Swedish words the woman knew was “pensionable income” (a social security term), and before she was hired by this company, this was something “she could only dream of”. Again, this form of marketing signals to prospective customers that the domestic services sector is a vital part in the integration and second language learning of new Swedes.

In sum, the contemporary data comprise examples of how references to language competence and opportunities for language learning are used in strategic communication to create positive values for marketing purposes. References to language are most often found in a section of the company website, which demarcates the qualities of this particular company, under headlines such as “This is why customers choose us”. In this sense, language becomes a commodified resource (Heller and Duchêne 2012) used as a quality benchmark for the company in at least two ways. First, some instances of marketing highlight the linguistic skills of the company’s workforce. Second, some companies allude to the language learning opportunities that their workforce face as a result of working for this company. Thus, language learning becomes a metalinguistic product, not so much for indexing the linguistic qualities of the employee, but rather indexing the moral qualities of the employer and, in effect, of the customer.

7 Concluding remarks

While sociolinguistic aspects of transnational migration and domestic work have mainly been studied from a contemporary perspective (e.g. Lorente 2018; Lutz 2011), this study has offered a historically grounded understanding of how language becomes part of the infrastructure of migratory domestic work. The study has investigated how references to language competence and language learning are turned into metalinguistic products that are being used as discursive strategies to navigate, and possibly avoid, upset in the morally loaded space of migration and domestic labor.

The study shows that marketing for domestic work often adapts discursive strategies to construct positive social images of trustworthiness, arguably in response to the destabilization of the line between private and occupational space associated with this kind of service. Moreover, the advertisements in both datasets materialize social relations between the domestic worker and the buyer of these services. In this regard, there is a significant difference between the datasets concerning how the provision of domestic services is arranged: by live-in maids or by cleaners contracted by the hour through agencies, mostly performing the work when no one is at home. This difference can be an additional explanation of why the historical advertisements are more centered on the individual who would perform the services, whereas contemporary advertisements rather showcase the company that is selling the services. The discursive strategies used are also contingent upon factors such as who the author is, whom they perceive as prospective addressees, and how they perceive the tensions in the market. Thus, in Bourdieu’s words, these discursive strategies point to how “the conditions of reception envisaged are part of the conditions of production” (1991: 76). Thus, when authors of advertisements include statements about the linguistic aspirations or skills of the domestic worker, this should be interpreted in relation to the “envisaged reception” that the advertisement would have.

In the personal, historical advertisements, many job seekers foregrounded their willingness to learn Swedish. Doing this, they tacitly align to an established sociolinguistic order and pre-empt potential concerns about their lack of linguistic proficiency. Additionally, foregrounding their language learning aspirations can also be a way to create a positive social image around themselves and their second language speakership. This discursive strategy can be interpreted in the light of the unmet demand for domestic services during the mid-twentieth century that made language requirements difficult, and the mutual dependency between the employer and employee. Finally, since live-in maids have closer interactions with their employer, it is a realistic ambition for the employee to eventually learn the language used in the household and in society at large.

In the historical data, it is notable how the individuals who advertised for a placement fronted their aspiration to learn Swedish as a reason for migrating to Sweden, rather than expressing more profane needs such as needing an income or, in some cases, fleeing war and economic despair. One could assume that this is because such profane needs have too many negative associations, but in that case, the logical inference is that the process whereby immigrants learn the majority language is, at least to a larger extent, loaded with positive values. By contrast, the contemporary data show how the linguistic competence of the staff is portrayed as a requirement from the company, pointing to a perception of a linguistic market where language competence is seen as a minimum requirement and as a prerequisite for work (Lorente 2012).

In the contemporary marketing from cleaning companies, the language competence of the workforce is turned into a mark of quality, and opportunities for language learning is showcased in the company’s ethical profiling. This strategy can be interpreted in the light of how immigration in general, and adult migrants’ language competence in particular, are constructed as a threat to social cohesion and security in public debates (e.g. Kahn 2019; Kraft 2019). Foregrounding the importance of language competence and language learning becomes a strategy to meet and align with such a discourse (Rydell 2015). Furthermore, with respect to previous mediatized debates about various moral aspects concerning commodified domestic services, the ethical profiling can be understood as semiotic behavior aiming to control the interpretation of the social image of the company. In this sense, the public self-promotion by the cleaning agencies becomes a discursive strategy with which to navigate the space of upset by alluding to its potential to solve personal as well as societal problems. By creating “a link between a brand and a consumer’s self-concept” (Escalas 2008: 168), corporate narratives construct an affective link to the services they market. As the company promotes itself as responsible when offering employees fair working conditions and opportunities for language learning, as well as facilitating their integration into Swedish society, the company simultaneously positions the prospective customer as morally upright.

The Bourdieusian framework proposes that individuals make assessments about the linguistic market before placing their goods – that is, before using their language, or presenting their ways of using language, in a certain way (Bourdieu 1991; Salö 2015). If this is correct, that logic should also explain why those who advertise domestic work choose to allude to the fact that many domestic workers in Sweden speak Swedish as a second language. According to this logic, the recurrent references to migration and second language speakership in advertisements for domestic work should be understood as reflecting the assumption that the addressees of the advertisements – prospective buyers of domestic services – will find the marketing more agreeable and trustworthy with such references than without them. Overall, domestic services carried out by immigrated labor is often envisioned as embodying a stratified society that is met with unease, historically as well as contemporarily. Against this backdrop, it seems that advertisers draw on different ideologies concerning the relation between language and work, that is, language competence as a requirement, or the workplace as a site for language learning, in order to charge migratory domestic labor with positive values. In conclusion, the study suggests that where migratory domestic work is a space of upset, referring to the second language learning of migrants is, it appears, a potential refuge.